While strong evidence suggests that the medial temporal lobe (MTL) is essential for learning and retaining new information for facts and events and the striatum is important for acquiring new skills and habits, the nature of the specific interactions between these two brain regions remains poorly understood. The goal of this dual-PI proposal is to take advantage of the precise spatial and temporal resolution of behavioral neurophysiology studies in animal model systems (Suzuki) together with the broad activation monitoring and flexible behavioral manipulation available in BOLD fMRI studies in humans (Stark) to characterize the specific contributions and interactions between the MTL and the striatum during a conditional motor associative learning task known to be dependent on both areas. In Aim 1, we will use the same task in both experimental animals and humans to assess the patterns and temporal dynamics of neural activity in the MTL and striatum during new conditional motor associative learning. Neurophysiology studies will include single unit tetrode recording, network correlation analyses and LFP analyses across both the MTL and the striatum. The BOLD fMRI studies will include characterization of functional connectivity between these areas. We will test the hypothesis that both the MTL and striatum signal learning during new conditional motor associative learning, but utilize distinct computational principles during the learning process such that the MTL associates random element together in memory while the role of the striatum includes motor- based or direction-based stimulus-response learning as well as a prominent role in signaling reward prediction error. We will also test the hypothesis that the role of the striatum in signaling reward prediction error interacts directly with the MTL defining a "declarative" portion of the striatum described in previous studies. In Aim 2, Stark will use various task manipulations hypothesized to make the associative learning task more dependent on either the MTL or the striatum to better characterize the unique contributions of these two different brain areas to associative learning. In Aim 3 Stark and Suzuki will conduct a detailed comparison of the pattern of single unit activity, LFP signals and spike-field coherence measured in animals to the pattern of BOLD fMRI signals and functional connectivity measured in humans to define the relationship between these different levels of analysis. Understanding the details of this relationship will be essential for ultimately translating experimental single cell findings in animals to our understanding of human brain function. Understanding the functional interactions between the MTL and striatum also has important implications for the development of treatments of a wide variety of disease states that affect these brain areas including Alzheimer's disease, attention deficit disorders, cognitive impairments present in aging, Parkinson's disease and Huntington's disease. PUBLIC HEALTH RELEVANCE: Alzheimer's disease, schizophrenia, developmental disorders and aging all involve impairments in learning and memory associated with damage to the medial temporal lobe while Parkinson's disease and Huntinton's disease involve damage to the striatum. Here we propose to use a combination of BOLD fMRI approaches in humans and single unit neurophysiological recording techniques in non-human primates to characterize the individual contributions and interactions of both medial temporal lobe areas and striatal areas important for new associative learning. Understanding the detailed relationship between BOLD fMRI signals in humans and single unit physiology signals in non-human primates will help us realize the potential of non-human primate model systems for understanding human cognition. This information will also serve as an important foundation for the development of treatments for disorders of memory, cognition and motor function that affect the medial temporal lobe and striatum.